Prosopagnosia, an inability to recognize faces, is a socially debilitating disorder. Therefore, training programs that investigate the ability to improv face recognition are essential for both treating the disorder and gaining a better understanding of the face recognition system. However, published attempts at training face recognition have been few. The primary goal of this proposal is to create and evaluate a training program based on perceptual learning principles that target the ability of prosopagnosic subjects to perceive facial structure. Our central hypothesis is that this training will improve behavioral and neural responses to faces in prosopagnosia. We will study a relatively large cohort and collect a comprehensive profile for each participant, including behavioral and fMRI data. These data will address our secondary goal, which is to determine the necessary conditions for improvement: that is, which functional types of prosopagnosia improve, and what structures in the brain may be critical for successful training. Specific Aims: The specific aims of the project are to determine the intervention's effect on acquired and developmental prosopagnosic's behavioral response to faces (aims 1 and 3). Furthermore, we hope to evaluate the effect that the intervention has on prosopagnosic's neural response to faces (aims 2 and 4). The accomplishments of these aims will allow us to, for the first time, directly compare acquired and developmental prosopagnosia as well as clarify the functional and structural determinants of benefit from training. Method: Pre-intervention baseline assessments will include measures of face processing and neural activation. Participants will complete approximately 30 minutes of training, 3 times per week. This training will utilize an adaptive staircase method in which participants are gradually exposed to more similar pairs of faces as performance improves. With each phase of the training program, a greater degree of variance in expression and angle of the face will be introduced to increase the possibility that training effects will generalize to novel post-training stimuli and real-world situations. Pre- and post-training assessments will be compared to examine improvements in face processing skill and changes in neural activation. In addition, group data will be compared to age-matched controls. Health Relevance: In addition to advancing our understanding of the cortical face network and the plasticity of this system, the proposed work will provide an online training program for adults with prosopagnosia that will be widely available. Postdoctoral Training program: As part of the training program, the applicant will learn new skills in functional neuroimaging, new approaches and perspectives in neuro-ophthalmology, and will have the opportunity to extensively examine a large cohort of adults with prosopagnosia. *Changes from the first application are indicated in bold.